A. General
Modern medical practice typically requires routine clinical tests of sera and urine for biological analytes such as cholesterol, enzymes (such as creatine kinase, lactate dehydrogenase, acid phosphatase, alkaline phosphatase, amylase, etc.), immunoglobulins, as well as other substances.
More specific (and, typically, more time-consuming) diagnostic tests are also performed in addition to routine tests. For example, the detection of certain isozymes of acid phosphatase is used clinically as an indicator of prostatic cancer as well as various leukemias. Levels of certain isozymes of alkaline phosphatase detected in a blood or serum sample serve as an indicator of bone and liver metabolic activity. Levels of pancreatic specific amylase are an indicator for pancreatitis. Serum levels of the MB isozyme of creatine kinase (CRMB), as well as levels of isozymes of lactate dehydrogenase, are indicators of myocardial infarction (Noel, S. et al. "Enzymes" in Clinical Chemistry (Kaplan, L. and Pesce, A., eds. The C. V. Mosby Company, St. Louis, Mich.); pp. 454-483 (1989)). Similarly, the detection of cholesterol, in specific lipoprotein classes, is used in the determination of coronary heart disease risk. (Russel et al. "Lipids" in Clinical Chemistry (Kaplan, L. and Pesch, A., eds. The C. V. Mosby Company, St. Louis, Mich.); pp. 968-1004 (1989)). These more specialized tests are often directed to a specific class of analyte that is already the subject of routine tests.
The efficacy of assays for analytes in a biological fluid sample can be reduced due to the presence of substances which interfere with the assay (Kaplan, L. and Pesce, A., "Interferences in Chemical Analysis" in Clinical Chemistry (Kaplan, L. and Pesce, A., eds., The C. V. Mosby Company, St. Louis, Mich.); pp. 808-819 (1989)). For example, compounds such as hemoglobin or bilirubin, which have a strong visible absorbance, can interfere with a spectrophotometric assay for an analyte. Kaplan and Pesce, id.
Clinical testing, in the case of both routine and more specialized tests, demands strict adherence to carefully developed quality assurance and quality control procedures in order to assure accuracy and to minimize variability of test results. Concerns over variability and inaccuracy of test results have in fact led to further regulation of clinical laboratories by the Health Care Financing Administration of the U.S. Department of Health and Human Services. 53 Federal Register 29590-29632 (Friday, Aug. 5, 1988) (proposed amendments to 42 CFR part 74 et seq.); 55 Federal Register 9538-9610 (Wednesday, Mar. 14, 1990) (revision of laboratory regulations, final rule with request for comments). These new regulations impose additional burdens on clinical testing laboratories. Such laboratories thus have a need for testing procedures that can be readily verified for adherence to quality control standards. The more specialized tests (as opposed to the routine tests) may readily permit verification, but the inherently sophisticated nature of these tests requires mastery by the laboratory technician of a set of testing protocols entirely different from those used in connection with routine tests. The result is that quality control of such specialized tests typically requires more extensive laboratory procedures and training of laboratory personnel.
An additional complication is posed in the interpretation of test results, even assuming that there is good quality control from one test run to another. For example, because of the important diagnostic information gained from cholesterol results and the need to eliminate interlaboratory variability, uniform cholesterol cutpoints based on national population studies have been adopted. Additionally, a national reference system for cholesterol has been developed so that cholesterol measurements are standardized and values are therefore traceable to the National Reference System for Cholesterol. Due to the absence of accepted National Reference Systems for triglycerides, lipoproteins, and apolipoproteins, much remains to be done in the elimination of interlaboratory variability associated with these lipid related tests. Presently, these tests and other specialized tests for cholesterol may not be directly related to the National Reference System for Cholesterol.
With this discussion as background, the remainder of this Background Art section discusses cholesterol determination, as an example of the state of the art in the detection of analytes in biological fluids. The prior art known to the inventors lacks an assay, for cholesterol in specific lipoprotein classes, that is simultaneously (i) easily interpretable from an epidemiological point of view; (ii) easily, quickly and inexpensively implemented, and (iii) universally applicable to all routine clinical chemistry testing systems. Indeed, the inventors are unaware of any assays, for specific classes of an analyte that are the subject of the routine tests described above, that meets these two criteria.
B. Cholesterol
Biochemical Background
Triglycerides and cholesterol are transported in the blood via lipoprotein particles. Abnormalities in these lipoproteins, either inherited, environmentally contributed, or a combination of both, lead to a variety of disorders including a predisposition to premature coronary heart disease (CHD) and atherosclerosis (N.I.H. Publication Number 88-2925 (1988); and Schaefer and Levy, The New England Journal of Medicine 312:1300-1310 (1985)). The underlying cellular and genetic mechanisms of many of the disease states have been intensively and elegantly explored in the preceding 30 years (Brown and Goldstein, Science 232:34-47 (1986), and Lusis, J. Lipid Research 29:397-428 (1988)).
The chemistry, biosynthesis, function, metabolism, cell biology, and molecular genetics of lipoprotein particles have been extensively reviewed (Segrest, Jere P., and Albers, John J., editors, 128 Methods in Enzymology (1986) and Albers, John J., and Segrest, Jere P., editors, 129 Methods in Enzymology (1986)).
Lipoprotein particles are divided into four major classes based on their density, composition, and electrophoretic mobility: The classes are chylomicrons, very low density lipoproteins (VLDL), low density lipoproteins (LDL) and high density lipoproteins (HDL). LDL and HDL particles may be further subdivided on the basis of density. The lipoprotein particles are composed of triglycerides, cholesterol, fatty acids esters of cholesterol, phospholipid and protein. The varying ratios of protein to lipid, in different lipoprotein classes, account for the physical differences by which these particles can be fractionated by density gradient centrifugation.
The protein components, known as apolipoproteins, are responsible for a variety of cellular functions. Increased levels of LDL cholesterol and decreased levels of HDL cholesterol have been shown to be risk factors for CHD. Consequently, clinical diagnostic assays for cholesterol content in the major lipoprotein classes are performed extensively and a large body of statistical data on the normal ranges for these classes is available standardized by the Centers for Disease Control (McNamara and Schaefer, Clinica Chimica Acta 166:1-8 (1987)).
Cholesterol Determination
In the clinical laboratory, the following assays are performed routinely to characterize the lipid and cholesterol profile of a plasma or serum sample: (i) Triglycerides are determined using the enzyme lipase(s) plus enzymes linked to a color indicator system, (ii) Total cholesterol is determined enzymatically using cholesterol esterase, cholesterol oxidase and other enzymes and reagents which translate the oxidation of cholesterol into a detectable color change, and (iii) HDL cholesterol is determined enzymatically as for (ii) above in the supernatant of a sample following selective precipitation of the VLDL and LDL fractions using a mixture of polyanions, e.g. sulfated polysaccharides, or phosphotungstate and divalent cations (Burstein et. al., J. Lipid Research 11:583 (1970); and Mulder et. al. Clinica Chimica Acta 143:29-35 (1987)). VLDL and LDL cholesterol (VLDL.C, LDL.C) are measured indirectly using the Friedewald equation (Friedewald et. al., Clin. Chem. 18:499-502 (1972)):
Total cholesterol=HDL.C+VLDL.C+LDL.C PA1 LDL.C=Total.C-(VLDL.C+HDL.C) PA1 LDL.C=Total.C-(Triglycerides/5+HDL.C)
The equation assumes (i) that no chylomicrons are present, for example, in a blood sample from a fasting patient, and (ii) that there is a constant relationship between cholesterol and triglycerides: This is known to be untrue in hypertriglyceridemic conditions (Cohn et. al., Clinical Chemistry 34:2456-2459 (1988); and Rao et. al., Clinical Chemistry 34:2532-2534 (1988)).
Thus the above analytical procedures suffer from several disadvantages. (i) The VLDL.C and LDL.C are not measured directly but rather are estimated using a formula. (ii) The Friedewald formula is known to be imprecise under conditions of clinical relevance i.e. elevated triglyceride levels (&gt;400 mg/100 ml) (Cohn et. al., Clinical Chemistry 34:2456-2459 (1988); and Rao et. al., Clinical Chemistry 34:2532-2534 (1988)). (iii) HDL.C determination relies on the selective precipitation of VLDL and LDL particles by a polyanion, or by phosphotungstate, plus divalent cations, with subsequent total cholesterol measurement of the separated supernatant. In general, cholesterol detection in specific lipoprotein classes lacks a standardized reference system.
Mulder et. al. (Clinical Chimica Acta 143:29-35 (1987)) report on the direct measurement of cholesterol in redissolved LDL precipitates but such measurements are not performed in routine diagnostic surveys.
More recent efforts toward separation and quantitation of lipoprotein classes have utilized antibodies, either polyclonal or monoclonal, directed against apolipoproteins which are specific to distinct, clinically relevant lipoprotein particles (Tikkanen et. al., J. Lipid Research 24:1494-1498 (1983); and Ordovas et. al. J. Lipid Research 28:1216-1224 (1987)).
In research laboratories a variety of immuno-based analytical techniques have been employed to quantitate lipoproteins, including radial immunodiffusion, radioimmunoassay and electroimmunoassay, but these techniques are too cumbersome to be employed in a clinical diagnostic setting where large numbers of samples must be handled rapidly. This disadvantage may be addressed by using an enzyme-linked immunoabsorbant assay (ELISA), and this is an area of active investigation (Ordovas et. al., J. Lipid Research 28:1216-1224 (1987)).
However, there is a further disadvantage which some of these immuno-based techniques, including ELISA suffer, and that is that these procedures quantitate an epitope associated with specific lipoprotein classes--they do not measure cholesterol levels. The significance of this situation is that there must be a very large number of samples analyzed by an immuno-based procedure to establish its correlation with cholesterol values (measured enzymatically) and which are interpretable epidemiologically. Thus, ELISA-based tests require a long lead time to gain acceptance in the clinical diagnostic industry.
Methods which utilize immobilized antibodies to measure levels of substances in biological fluids are known. Longenecker (U.S. Pat. No. 4,302,536 (1981)) reported the determination of antigenic materials in biological fluids and cells by colorimetric immunoassay with an adduct of antibody and chromo-protein. Onishi and Ito (Eur. Pat. No. 327,918 (1989)) reported an immunoassay using the homogeneous competitive reaction between a target and labelled substance and a specific binder. Freytag and Ishikawa (U.S. Pat. No. 4,657,853 (1987)) reported a high sensitivity immunoassay using a polymeric enzyme-antibody conjugate. Nippon (Jap. Pat. No. 59226864 (1984)) reported an immunoassay in which levels of transforming growth factor (TGF) in a liquid sample are detected using an immobilized TGF antibody and an enzyme labelled TGF antibody. Gomez and Wicks (U.S. Pat. No. 4,353,982 (1982)) report an immunoassay for creatine kinase in blood serum using iodine-125 labelled antibody to precipitate immune complex mixtures.
Several groups have examined selective immunoprecipitation of specific lipoprotein classes followed by cholesterol quantitation in the lipoprotein class remaining in solution. Heuck et al. reported the use of antibodies to ApoB to precipitate LDL and VLDL followed by measuring cholesterol levels in the HDL left in the supernatant. Antibodies to apoAI and apoC were also used, to precipitate HDL and VLDL, followed by determination of cholesterol levels in the LDL left in the supernatant. (Heuck et al. Clin. Chem. 31:252-258 (1985)). Kerscher et al. reported the use of antibodies to HDL to precipitate HDL and VLDL, followed by centrifugation to separate the precipitate, followed by analysis of cholesterol levels, or other component levels, in the LDL in the supernatant (Kerscher et al. U.S. Pat. No. 4,746,605 (1988); Fed. Rep. Germany Patent No. P32 15 310 (1983); Kerscher et al. Clin. Bioch. 18:118-125 (1985)). Antibodies to both apoproteins and whole lipoproteins, including immobilized antibodies, have been used to immunoprecipitate lipoproteins followed by determination of the cholesterol content of the lipoprotein class remaining in solution (Ziegenhorn et al. Canadian Patent No. 1 211 707 (1986)). This reference, however, does not describe any specific structure or device on which the antibodies are immobilized.